4. How we can decide something is true
We have two tools
Observation-we have our
sense through which we
experience the world
Inferences- Ability to
make logical inferences
6. Inferences
• Deductive – we predict
what the observation
should be
• Inductive- we go from
specific to general
7.
8. Deductive inferences
• Positive reinforcement result in better
learning than the punishment.
• Rewards work better than punishment
• Math students who praised for their right
answer during the year will do better on the
final exam than those who are punished for
their wrong answer
9. • We go from general theory to specific theory
observation
• This is known as hypothetic deductive method
10. Inductive interferences
• In vienna general hospital 1840 that
woman giving birth were dying at
puerperal fever.
Note
• Puerperal-during or relating to the period
of about six weeks after child birth
during the mothers reproductive organ
to their original non pregnant condition.
11. Doctor Philip after observed medical
student performing vaginal examination
did so directly coming from the dissecting
room, rarely washing their hands in
between with the observation that a
colleague who accidentally cut his finger
while dissecting a corpse dies of malady [
illness/ infection]
Inductive interferences
12. • He inferred the explanation that the cause of
death was the introduction of cadaverous
material into a wound .
• The practical consequence of that creative leap of
imagination was elimination of fever as a course
of child birth by requiring that physicians wash
their hands before doing a delivery.
• Epidemiologist generally use inductive
interference.
Inductive interferences
13. • Theories can be used to predict observation.
• The observations will not always be exactly as we
predict them, due to error and the inherent
variability of natural phenomena.
• If the observations are widely different from our
predictions we will have to abandon or modify
the theory.
How to do we test the extent of the discordance of
our prediction based on theory from the reality of
our observation.
The test is statistical or probabilistic test.
14. Deterministic model
A phenomenon may be principally based on
deterministic model.
Example
Boyle’s law for a fixed volume an increase in
temperature of a gas determines that there is an
increase in pressure.
Each time the law is tested the same result occur.
The only variability lies in the error of
measurement. Many phenomena in physics and
chemistry are of such a nature.
15. Probabilistic model
• Various states of phenomenon occur with
certain probabilities.
• In biology/psychology/ medicine where
phenomena are influenced by many factors
that in themselves are variable and by other
factors that are unidentifiable, the models are
often probabilistic.
16. • The model is principally probabilistic, statistical
techniques are needed to increase scientific
knowledge.
• The presence of variation require the use of
statistical analysis.
• When there is a little variation with respect to a
phenomenon, much more weight is given to a
small amount of evidence.
• Example: Pancreatic cancer appears to be
invariably fatal disease.
17. • A drug that indisputably cured a few patient of
pancreatic cancer we would give a lot of weight
to the evidence that the drug represented. [ The
disease were more variable]
• Vitamin C cures common cold.
• We need to demonstrate its effect in many
patient and need to use statistical method to do
so.
• Human beings are quiet variable with respect to
cold.
18. Statistical methods are objective method by
which group trends are abstracted from
observation on many separate individuals.
A simple concept of statistics is the calculation
of averages percentages and so on.
The presentation of the data in tables and
charts, such techniques is important to describe
the population in the study.
19. • Average of blood pressure of women college
student.
• Measure the blood pressure of every single
member of this population we would not have to
infer anything.
• We would simply average all the numbers we
obtained.
• In practice we take a sample of students [
properly selected] on the basis of the data we
obtain from the sample.
• We infer the mean of whole population is likely to
be.
20. • In statistical reasoning then we make
inductive interferences from the particular to
general. Thus the statistics may be said to be
the technology of the scientific method.
21. Design of studies
• The testing of hypotheses must be done by
making controlled observation, free of systematic
bias.
• Statistical techniques to be valid , must be
applied to data obtained from well designed
studies otherwise solid knowledge is not a
advanced.
Study type
• Observational
• Experimental
22. • Observational-
• Nature determines who is exposed to the
factor of interest and who is not exposed.
• Experimental –
• The investigator determines who is exposed
these may causation.
23.
24. Cross sectional studies
• Cross sectional study measurement are taken
at one point in time
• Example
• Cross sectional study of high blood pressure
and heart diseases at same time.
• It is useful in showing association in providing
easy clue to etiology.
25. Case control studies
• Investigator starts with lung cancer cases and with
control through examination of the record or through
interviews determines the presence or absence of
factors in which he or she is interested [ smoking].
• It also referred as retrospective study.
• The data on the factor of interest are collected
retrospectively and thus may be subject to various
inaccuracies.
• It is useful in rare disease or condition or when the
disease takes a very long time to become manifest.
26. Prospective / cohort study
• The investigator starts with a cohort of non-
diseased person with that factor [ i.e those
who smoke ] and person without that factor [
non smoker ] and goes forward into some
future time to determine the frequency of
development of the disease in the two group.
• It is also known as longitudinal study.
• It is very useful for stronger evidence of
casuality and less subject to bias.
27. • Prospective studies provide stronger evidence of
casuality than retrospective studies but more
difficult , more costly and sometimes impossible
to conduct.
• The disease under study takes decades to
develop or if it is very rare.
• In health field, an experimental study to test an
intervention of some sort is called a clinical trial.
• Clinical trial prospective, experimental studies
that provide the most vigorous evidence of
casuality
28. Data
Discrete variables
• It assume certain fixed numerical values
• Ex.Sex is discrete variable code male 1 and
female-2
Continuous variables
It assume infinite number of values between
two fixed point
Weight is a continuous variable.